Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1530598 > unrolled thread

Re: [PATCH v3 3/3] zram: support BDI_CAP_STABLE_WRITES

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v3 3/3] zram: support BDI_CAP_STABLE_WRITES Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2016-11-26 07:40 +0100
    Re: [PATCH v3 3/3] zram: support BDI_CAP_STABLE_WRITES Minchan Kim <minchan@kernel.org> - 2016-11-26 15:50 +0100
      Re: [PATCH v3 3/3] zram: support BDI_CAP_STABLE_WRITES Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2016-11-27 14:10 +0100

#1530598 — Re: [PATCH v3 3/3] zram: support BDI_CAP_STABLE_WRITES

Hello Minchan,

On (11/25/16 17:35), Minchan Kim wrote:
[..]
> +static void zram_revalidate_disk(struct zram *zram)
> +{
> +	revalidate_disk(zram->disk);
> +	zram->disk->queue->backing_dev_info.capabilities |=
> +		BDI_CAP_STABLE_WRITES;
> +}
> +
>  /*
>   * Check if request is within bounds and aligned on zram logical blocks.
>   */
> @@ -1094,7 +1102,7 @@ static ssize_t disksize_store(struct device *dev,
>  	zram->comp = comp;
>  	zram->disksize = disksize;
>  	set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
> -	revalidate_disk(zram->disk);
> +	zram_revalidate_disk(zram);
>  	up_write(&zram->init_lock);
>  
>  	return len;
> @@ -1142,7 +1150,7 @@ static ssize_t reset_store(struct device *dev,
>  	/* Make sure all the pending I/O are finished */
>  	fsync_bdev(bdev);
>  	zram_reset_device(zram);
> -	revalidate_disk(zram->disk);
> +	zram_revalidate_disk(zram);
>  	bdput(bdev);
>  
>  	mutex_lock(&bdev->bd_mutex);

why not set it just once, when we allocate queue/disk and configure both
of them:  in zram_add()

	queue->backing_dev_info.capabilities |= BDI_CAP_CGROUP_WRITEBACK;

	-ss

[toc] | [next] | [standalone]

#1530674

Hi Sergey,

On Sat, Nov 26, 2016 at 03:37:01PM +0900, Sergey Senozhatsky wrote:
> Hello Minchan,
> 
> On (11/25/16 17:35), Minchan Kim wrote:
> [..]
> > +static void zram_revalidate_disk(struct zram *zram)
> > +{
> > +	revalidate_disk(zram->disk);
> > +	zram->disk->queue->backing_dev_info.capabilities |=
> > +		BDI_CAP_STABLE_WRITES;
> > +}
> > +
> >  /*
> >   * Check if request is within bounds and aligned on zram logical blocks.
> >   */
> > @@ -1094,7 +1102,7 @@ static ssize_t disksize_store(struct device *dev,
> >  	zram->comp = comp;
> >  	zram->disksize = disksize;
> >  	set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
> > -	revalidate_disk(zram->disk);
> > +	zram_revalidate_disk(zram);
> >  	up_write(&zram->init_lock);
> >  
> >  	return len;
> > @@ -1142,7 +1150,7 @@ static ssize_t reset_store(struct device *dev,
> >  	/* Make sure all the pending I/O are finished */
> >  	fsync_bdev(bdev);
> >  	zram_reset_device(zram);
> > -	revalidate_disk(zram->disk);
> > +	zram_revalidate_disk(zram);
> >  	bdput(bdev);
> >  
> >  	mutex_lock(&bdev->bd_mutex);
> 
> why not set it just once, when we allocate queue/disk and configure both
> of them:  in zram_add()

I should have mentioned the reason.
The revalidate_disk reset the BDI_CAP_STABLE_WRITES.

Thanks.

[toc] | [prev] | [next] | [standalone]

#1530873

On (11/26/16 23:41), Minchan Kim wrote:
[..]
> > >  	mutex_lock(&bdev->bd_mutex);
> > 
> > why not set it just once, when we allocate queue/disk and configure both
> > of them:  in zram_add()
> 
> I should have mentioned the reason.
> The revalidate_disk reset the BDI_CAP_STABLE_WRITES.

aha. either sets or clears it in blk_integrity_revalidate(),
now I see it.

	-ss

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web